AI-Enhanced Coral Bleaching Prediction: Bridging Gaps in Existing Monitoring Platforms
International Journal of Advanced Research in Science Communication and Technology · 13.05.2026 · 287
DOI: 10.48175/ijarsct-35339 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 13.05.2026. Verfasst von Srishti Aggarwal und 3 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben.
Abstract (Originalsprache)
Coral reef ecosystems face an accelerating threat from thermally driven bleaching events that now recur faster than reefs can recover between episodes. The core operational challenge is not data scarcity but the absence of field-ready tools capable of converting raw photographic imagery into actionable, colony-level risk intelligence without specialist hardware or technical expertise. This paper presents CoralGuard AI, a browser-deployable reef monitoring application that couples a fine-tuned YOLOv8-Nano object detection model with an interactive Streamlit dashboard. The system accepts JPEG or PNG field imagery, performs single-pass inference, and returns annotated bounding-box overlays alongside a computed ecosystem mortality risk score derived from the ratio of bleached to total detected coral colonies. Concurrent environmental context , sea-surface temperature, pH, UV index, and dissolved oxygen , is rendered simultaneously in a live telemetry panel. Evaluated on a varied corpus of reef survey photographs, the pipeline achieves a mAP50 of 0.83 and correctly assigns alert severity in 14 of 17 held-out test frames. The full deployment stack (Ultralytics, Streamlit, OpenCV, Pillow, NumPy) operates on standard consumer CPU hardware without GPU support and requires no local installation or command-line interaction, directly addressing the field-deployment gap that limits practical uptake of automated reef assessment tools.
Zitieren
APA
Aggarwal, S., Jain, A., Yadav, T., & Goswami, S. (2026). AI-Enhanced Coral Bleaching Prediction: Bridging Gaps in Existing Monitoring Platforms. International Journal of Advanced Research in Science Communication and Technology, 287. https://doi.org/10.48175/ijarsct-35339
Vancouver
Aggarwal S, Jain A, Yadav T, Goswami S. AI-Enhanced Coral Bleaching Prediction: Bridging Gaps in Existing Monitoring Platforms. International Journal of Advanced Research in Science Communication and Technology. 2026:287. doi:10.48175/ijarsct-35339
BibTeX
@article{Aggarwal2026,
author = {Aggarwal, Srishti and Jain, Ambika and Yadav, Tanishq and Goswami, Shristy},
title = {AI-Enhanced Coral Bleaching Prediction: Bridging Gaps in Existing Monitoring Platforms},
journal = {International Journal of Advanced Research in Science Communication and Technology},
year = {2026},
pages = {287},
doi = {10.48175/ijarsct-35339}
}
EUVIMED